BKARINSO.RVW   990715

"Architects of the Information Society", Simson L. Garfinkel, 1999,
0-262-07196-7, U$20.00
%A   Simson L. Garfinkel simsong@vineyard.net simsong@aol.com
%C   55 Hayward Street, Cambridge, MA   02142-1399
%D   1999
%E   Hal Abelson
%G   0-262-07196-7
%I   MIT Press
%O   U$20.00 +1-800-356-0343 fax: +1-617-625-6660 www-mitpress.mit.edu
%P   72 p.
%T   "Architects of the Information Society"

Although concerned with MIT's seminal Project MAC and the subsequent
Laboratory for Computer Science (LCS), this book is not a history as
such.  Instead, it comprises three essays describing the development
of multiuser computing, the development of computer networking, and
the resulting impact of computer networking on society and business.

The concept of multiple access computing, multiuser computing, or time
sharing involves a number of original technologies such as interrupts,
security, virtual memory.  In addition, the project spawned research
in areas as diverse as operating system design, processor independent
architectures, and software portability.  While these topics are
mentioned, they are not covered.  Instead, the article concentrates on
personal or personnel matters: who did what, or went where, or hated
whom.  There is interesting social history, but little technical
record, and this lack may disappoint the technical audience.

A similar style characterizes chapter two.  In detailing the growth of
local networking, for example, the work only tangentially touches on
the unexpected desire for connection of the ARPANET's original IMPs
(Interface Message Processors) to multiple machines at a single site,
and the IMP redesign that was required.  In fact, the coverage of many
important aspects of networking, such as the development of the
ARPANET and Ethernet, are presented in very sketchy terms.

Chapter three is actually a grab bag of projects that have had LCS
involvement over the years, such as voice interface processing,
encryption, the World Wide Web, and artificial intelligence for
medical work.  In this section the book devolves almost completely
into anecdote, with very little structure or analysis to the stories.

As noted, the book is not a history.  However, even as an anniversary
celebration, the shortcomings are frustrating.  I can fully sympathize
with the feeling that modern companies unfairly take credit for
technologies invented earlier, but LCS is not best promoted by
ignoring other centres and groups, as seems to happen at times.  While
a number of the stories presented are interesting and little known, it
is difficult to recommend this volume to other than die-hard computer
history buffs.

copyright Robert M. Slade, 1999   BKARINSO.RVW   990715